Selected work
This collection prioritizes real client projects and the strongest technical work behind the services I offer. Use the filters to explore by type.
A real client driving-school platform focused on presenting services clearly and supporting the learner journey through a modern web experience.
A real client web project delivered as an end-to-end business solution, covering the customer-facing experience and the technical work required to prepare it for production.
A local-first desktop product that gives developers and AI agents durable, evidence-backed project memory across code, tasks, decisions, claims, validation, and AI handoffs.
An end-to-end logistics ML pipeline for freight-rate prediction using chronological validation, CatBoost, reproducible training, release checks, automated testing, and production-style inference artifacts.
A full-stack scholarship platform that helps students discover and work with scholarship opportunities through a modern web experience backed by API, authentication, database, and AI capabilities.
A privacy-conscious local-first computer-vision attendance MVP covering employee enrollment, face validation and matching, attendance rules, SQLite persistence, filtering, export, testing, and CI.
A reproducible educational ML classification pipeline with XGBoost, class-imbalance handling, SHAP explanations, automated tests, CI, and an interactive Streamlit demo.
A leakage-aware sports analytics pipeline that combines chronological Elo, rolling team form, probabilistic model selection, calibration evaluation, explainability, and automated tests.
A real-time computer-vision application that turns hand gestures into drawing actions using MediaPipe landmarks, testable gesture logic, OpenCV rendering, and automated quality checks.
An end-to-end Ames Housing regression project with reusable preprocessing, cross-validated model selection, artifact persistence, automated testing, and a Streamlit prediction interface.
Fruit Ninja — Hand Tracking
A real-time computer-vision game using MediaPipe hand tracking, velocity-based slicing, collision logic, physics, scoring, combos, lives, and automated core-logic tests.
